promptfoo

license

MIT(OSI)

source

public(GitHub)

core features withheld

no

MIT-licensed and fully public, and MIT is an OSI license, so this is open source. OpenAI now owns the project, which does not change the license the code ships under.

• https://api.github.com/repos/promptfoo/promptfoo/license recorded 2026-08-13

  GitHub's license endpoint reports spdx_id MIT for promptfoo/promptfoo and returns the MIT text as the LICENSE body.

• https://github.com/promptfoo/promptfoo recorded 2026-06-04

  MIT license; 'remains open source and MIT licensed' under OpenAI; v0.121.14

• https://raw.githubusercontent.com/promptfoo/promptfoo/main/README.md recorded 2026-08-13

  README states 'Promptfoo is now part of OpenAI. Promptfoo remains open source and MIT licensed' and lists 'Open source - MIT licensed, with an active community' among its properties. The CLI installs and runs from npm with no account, and nothing in the README offers a paid tier, an enterprise edition or a license key.

npm is the right registry to read from, since promptfoo is distributed npm-first and its downloads there are the primary usage signal, alongside GitHub stars. A PyPI package under the same name is genuinely this project's, but it is not the product's primary distribution, so reading from it would substitute a minority channel for promptfoo's real distribution and understate its reach.

• https://api.npmjs.org/downloads/point/last-month/promptfoo recorded 2026-08-13

  2,277,977 npm downloads for 2026-07-11 to 2026-08-09

• https://github.com/promptfoo/promptfoo recorded 2026-06-04

  21.9k stars; '10M+ users in production' claim

• https://raw.githubusercontent.com/promptfoo/promptfoo/main/README.md recorded 2026-08-13

  The repo carries 24.2k stars and the README positions the tool as used by OpenAI and Anthropic. No standalone user count is published beyond that.

Broad eval coverage for LLM applications, including prompt and model testing, RAG pipeline evaluation, side-by-side model comparison, CI/CD automation, and red teaming with vulnerability scanning. It sits below academic-benchmark standardization references such as lm-evaluation-harness on raw benchmark coverage.

• https://github.com/promptfoo/promptfoo recorded 2026-06-04

  eval + red-teaming + model comparison + CI/CD features

• https://raw.githubusercontent.com/promptfoo/promptfoo/main/README.md recorded 2026-09-29

  README describes "a CLI and library for evaluating and red-teaming LLM-based apps" and lists "automated evaluations", "red teaming", "side-by-side (OpenAI, Anthropic, Azure, Bedrock, Ollama" model comparison, and "CI/CD" automation, plus "Fast, with features like live reload and caching" and a "code scanning" review of pull requests.

• https://www.promptfoo.dev/docs/getting-started/ recorded 2026-09-29

  Getting started guide says "The YAML configuration format runs each prompt through a series of test cases" and uses a promptfooconfig.yaml file.

• https://www.promptfoo.dev/docs/intro/ recorded 2026-09-29

  Docs intro lists "Test and evaluate your prompts, models, and RAG pipelines" and states "promptfoo is fast, with quality-of-life features like live reloads and caching."

• https://www.promptfoo.dev/docs/usage/node-package/ recorded 2026-09-29

  Node package page says "Integrate LLM testing into Node.js apps" with promptfoo's evaluate() function.